Accepted Manuscript Title: Wheat genome specific granule-bound starch synthase I differentially influence grain starch synthesis Author: Geetika Ahuja Sarita Jaiswal Pierre Hucl Ravindra N. Chibbar PII: DOI: Reference:
S0144-8617(14)00761-9 http://dx.doi.org/doi:10.1016/j.carbpol.2014.08.004 CARP 9131
To appear in: Received date: Revised date: Accepted date:
28-3-2014 1-8-2014 4-8-2014
Please cite this article as: Ahuja, G.,Wheat genome specific granule-bound starch synthase I differentially influence grain starch synthesis, Carbohydrate Polymers (2014), http://dx.doi.org/10.1016/j.carbpol.2014.08.004 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Wheat genome specific granule-bound starch synthase I differentially influence grain
2
starch synthesis
3
Authors: Geetika Ahuja, Sarita Jaiswal, Pierre Hucl, Ravindra N. Chibbar
4
Affiliation address: Department of Plant Sciences & Crop Development Centre, College of
5
Agriculture and Bioresources, 51 Campus Drive, University of Saskatchewan, Saskatoon SK
6
S7N 5A8, Canada
7
Email addresses: [email protected], [email protected], [email protected],
8
[email protected]
9
Corresponding author: Professor Ravindra N. Chibbar, Department of Plant Sciences & Crop
10
Development Centre, College of Agriculture and Bioresources, 51 Campus Drive, University of
11
Saskatchewan, Saskatoon SK S7N 5A8, Canada
12
E-mail: [email protected]
13
Phone: +1-306-966-4969
14
Fax: +1-306 966-5015
Ac ce p
te
d
M
an
us
cr
ip t
1
1
Page 1 of 33
Abstract
16
Wheat grain development is a complex process and is characterized by changes in
17
physicochemical and structural properties of starch. The present study deals with endosperm
18
starch physicochemical properties and structure during development in different granule-bound
19
starch synthase I (GBSSI) null also known as waxy (Wx) genotypes. The study was conducted
20
with pure starch isolated from wheat grains at 3-30 days post anthesis (DPA), at three day
21
intervals. Amylose concentration increased throughout grain development in non-waxy (7.2% –
22
30.5%) and partial waxy genotypes (6.0%– 26.8%). Completely waxy genotype showed 7.0%
23
amylose at 3 and 6 DPA, which declined during development and reached non-detectable
24
quantities by 30 DPA. Amylopectin structure had a higher content of short chains at 3 DPA,
25
which decreased continuously until 12 DPA, after which there were only minor changes in
26
amylopectin chain length distribution. Similarly, the average degree of polymerization (DP)
27
increased from 3 DPA (12.3) to 12 DPA (15.0), and then did not differ significantly up to 30
28
DPA (15.0). This suggests the formation of basic amylopectin architecture in wheat by 12 DPA.
29
Wx-B and Wx-D affected amylopectin short chains mostly of DP 6-8 at 3 and 6 DPA. Wx-A
30
affected the same fraction of chains at 9 and 12 DPA, and Wx-D affected DP 18-25 chains from
31
18 to 30 DPA, suggesting differential effect of waxy isoproteins on amylopectin structure
32
formation.
33
Keywords: Granule-bound starch synthase I, waxy, grain development, amylopectin chain
34
length distribution.
35
Abbreviations1
Ac ce p
te
d
M
an
us
cr
ip t
15
1
DPA – Days post anthesis, DP – Degree of polymerization, GBSSI – Granule-bound starch synthase I, SEM –
Scanning electron microscopy
2
Page 2 of 33
36
1. Introduction
38
Biochemical properties of starch have been extensively utilized in the food industry for uses such
39
as thickeners for soups, sauces and dairy products, starch-based fat replacers to maintain texture,
40
flavor and mouth-feel of low- or no-fat products. The relative proportion and structure of starch
41
constituents, amylose and amylopectin, greatly influence its end-use. For instance, waxy wheat
42
starches possess exceptional freeze-thaw stability, water absorption capacity, peak viscosity, and
43
low gelatinization temperatures; hence they are useful in the baking industry (Abdel-Aal, Hucl,
44
Chibbar, Han, & Demeke, 2002). Amylopectin chain length distribution also affects
45
gelatinization, retrogradation, and pasting properties of starch. Short glucan chains reduce
46
gelatinization temperatures (Jobling, 2004), while very-long branch chains influence starch-
47
pasting properties (Jane et al., 1999); thus making them valuable in food industry applications.
48
Structurally, starch is a glucan homopolymer composed of one-quarter amylose (linear with few
49
branches) and three-quarters amylopectin (highly branched). Linear chains are formed by α-1,4
50
glucosidic linkages, while branches are formed by α-1,6 glucosidic linkages. The structures of
51
amylose and amylopectin, and their organization in the intact granule have been widely
52
discussed (Buléon, Colonna, Planchot, & Ball, 1998). Amylopectin shows a polymodal
53
distribution of chains, organized into clusters, giving the molecule its architectural specificity
54
(Hizukuri, 1986). Briefly, A-chains (DP 6-12) are the outermost chains, B-chains named B1, B2
55
and B3 depending on their degree of polymerization (DP 13-24 up to 50) support A-chains or
56
other B-chains. They are attached via α-1,6 glucosidic linkage, to the central C-chain, which
57
within a granule is oriented towards the centre or hilum of the granule (Manners, 1989). Packing
Ac ce p
te
d
M
an
us
cr
ip t
37
3
Page 3 of 33
of these chains, or rather the amylopectin molecule in the granule determines the basic structure
59
of the starch granule.
60
The first step in the synthesis of amylose and amylopectin is brought about by ADP-glucose
61
pyrophosphorylase (AGPase). Elongation and branching of amylopectin is a complex process
62
and it requires an array of enzymes viz., starch synthases (SS), starch branching enzymes (SBE)
63
and debranching enzymes (DBE) (Buléon et al., 1998). Synthesis of amylose, however, is
64
brought about solely by the enzyme granule-bound starch synthase I (GBSSI) or waxy protein.
65
Plants lacking GBSSI are termed ‘waxy’. Recently, GBSSI has also been postulated to have a
66
role in amylopectin synthesis (Ral et al., 2006). Incorporation of [14C] glucose from ADP [14C]
67
glucose by GBSSI in granules from different species has been reported to be mainly into the long
68
chain fraction of amylopectin (Denyer, Clarke, Hylton, Tatge, & Smith, 1996). However, the
69
precise contribution of GBSSI to amylopectin synthesis is still unclear.
70
Contrasting reports exist on the differences in amylopectin chain length distribution due to the
71
absence of waxy protein. In Chlamydomonas and wheat waxy mutants that lack GBSSI, the
72
amylopectin molecule has been shown to differ from that of their non-waxy counterparts (Delrue
73
et al., 1992; Li et al., 2007). In contrast, no obvious difference in the glucan chain-length
74
distribution profiles of amylopectin molecules between waxy and non-waxy starch within three
75
chain-length groups were reported (Yasui, Matsuki, Sasaki, & Yamamori, 1996). These studies
76
were conducted on mature grains. To better understand changes in amylopectin chain length
77
distribution profiles, some studies have been conducted at different developmental stages.
78
(Waduge, Xu, Bertoft & Seetharaman, 2013; Kalinga, Tetlow, Liu, Yada & Seetharaman, 2014).
79
It was reported that the content of A- and short B-glucan chains increased in the waxy, and
80
decreased in the non-waxy type of wheat during grain development. Also, the content of mid-
Ac ce p
te
d
M
an
us
cr
ip t
58
4
Page 4 of 33
length B-glucan chains decreased in both waxy and non-waxy starch during grain development.
82
It was also suggested that GBSSI elongates the A- or short B-glucan chains to produce long B-
83
glucan chains of amylopectin (Cao, Hu, & Wang, 2012).
84
Since wheat is an allohexaploid (2n = 6x = 42), it has three doses of GBSSI, one from each
85
genome, namely Wx-A, Wx-B and Wx-D. The lack of waxy protein from each genome has been
86
shown to affect the end-use of starch. Durum wheat, which possesses waxy protein from two
87
genomes, is the best candidate for pasta production (Sharma, Sissons, Rathjen, & Jenner, 2002).
88
Similarly, double null-waxy mutants of wheat result in better texture of noodles (Zhao et al.,
89
1998). Considering GBSSI potentially contributes to amylopectin synthesis or elongation, its
90
absence from each genome could affect the glucan chain length distribution of amylopectin
91
and/or organization of glucan chains within the granule, resulting in changes in the morphology
92
of starch granules during wheat grain development. Therefore, the present study focused on
93
studying starch characteristics during wheat grain development in non-waxy and waxy-null
94
genotypes.
95
2. Materials and Methods
96
2.1 Plant material
97
Near-isogenic partial-waxy and non-waxy genotypes including four lines each of Wx-A+B+D+,
98
A-B+D+, A+B-D+, A+B+D-, A-B-D+, A-B+D-, A+B-D- and A-B-D- were used for the developmental
99
study. Plants were grown in a growth chamber in pots containing Redi-earth (W. R. Grace & Co.
Ac ce p
te
d
M
an
us
cr
ip t
81
100
of Canada, ON, Canada) at 23°C / 16 h light (350 μmol m-2 s-2 PPFD) and 19 °C / 8 h dark, and
101
were fertilized every 3 weeks with a slow release fertilizer, Nutricote-14-14-14:N-P-K (Plant
102
Products Co. Ltd., Brampton ON, Canada). Spikes, in triplicate, were tagged at anthesis, 5
Page 5 of 33
collected and frozen in liquid N2 at selected days post anthesis (DPA), viz., 3, 6, 9, 12, 15, 18,
104
21, 24, 27 and 30 days, and stored at -80°C for further experiments. Spikes were collected in
105
mid-afternoon consistently for all developmental stages. Starch was purified from freeze-dried
106
seeds from the middle of the spikes using a modified method (Peng, Gao, Abdel-Aal, Hucl, &
107
Chibbar, 1999) involving cesium chloride density gradient centrifugation (Asare et al., 2011).
108
2.2 Starch granule morphology by scanning electron microscopy
109
Prior to imaging under scanning electron microscope (SEM), starch granules were gold coated to
110
make them electrically conductive. Starch granules were spread on, or attached to a metal stub
111
with double-sided carbon tape and gold coated using a sputter gold coater (S150B, Edwards,
112
Wilmington MA, USA). Granule morphology was observed using a scanning electron
113
microscope (SU6600 Hitachi High-Technologies Europe GmbH, Krefeld, Germany) at 5 kV.
114
2.3 Total starch concentration determination
115
Total starch concentration was determined using an American Association of Cereal Chemists
116
International (AACCI) approved method (AACC method 76-13.01), in which ground meal
117
samples were hydrolyzed to dextrins and finally to D-glucose using α-amylase and
118
amyloglucosidase respectively. Total starch concentration was determined as free glucose by
119
measuring the absorbance at 510 nm (Hucl & Chibbar, 1996) and calculated on a percent dry
120
weight basis (McCleary, Gibson, & Mugford, 1997).
121
2.4 Amylose concentration determination by HPSEC
122
High performance size exclusion chromatography (HPSEC) was used to determine the amylose
123
concentration (Demeke, Hucl, Abdel-Aal, Båga, & Chibbar, 1999). Gelatinized starch samples
Ac ce p
te
d
M
an
us
cr
ip t
103
6
Page 6 of 33
(1mg/mL) were debranched using isoamylase one unit for each mg of starch (1000U/mL
125
Megazyme International Ltd., Wicklow, Ireland), freeze-dried, suspended in dimethyl sulfoxide
126
(DMSO) and injected (40 μL) into the column (PL gel MiniMix 250 x 4.6 mm ID column,
127
Polymer Laboratories Inc., Amherst MA, USA) attached with Waters 410 differential
128
refractometer . (Asare et al., 2011). Starch samples, column, and detector were maintained at 40,
129
100, and 45°C, respectively. DMSO (99%) with 0.4 % Lithium bromide was used as an eluent at
130
a flow rate of 0.2 mL/min. Data was collected and analyzed using Empower 1154
131
chromatography software (Waters Corporation, Milford MA, USA).
132
2.5 Protein detection and immunoblotting
133
Proteins were extracted from the starch granules by suspending 10 mg starch in 300 μL
134
denaturing electrophoresis buffer (Demeke, Hucl, Nair, Nakamura, & Chibbar, 1997). GBSSI
135
was separated using polyacrylamide {15% (w/v) 1 mm thick gel} denaturing gel electrophoresis
136
(Protean II, Biorad, Hercules CA, USA) (Demeke, Hucl, Nair, Nakamura, & Chibbar, 1997).
137
Separated polypeptides were visualised by silver staining (Gromova & Celis, 2006). The gel-
138
electrophoresis separated polypeptides were transferred on a nitrocellulose membrane at 10 V for
139
4 h (Gao & Chibbar, 2000). The membrane with transferred polypeptides was processed and
140
incubated with primary antibodies raised against wheat GBSSI, SSI, SSII and SBE (Gao &
141
Chibbar, 2000; Peng, Hucl, & Chibbar, 2001). The antigen-antibody complex was then detected
142
with goat anti-rabbit alkaline phosphate conjugate using a 5-bromo-4-chloro3’-indolyphosphate
143
p-toluidine / nitro-blue tetrazolium chloride immunoscreening color development kit (Biorad,
144
Hercules CA, USA).
Ac ce p
te
d
M
an
us
cr
ip t
124
7
Page 7 of 33
2.6 Amylopectin chain length distribution by capillary electrophoresis
146
Chain length distribution of amylopectin molecules was determined by fluorophore assisted
147
capillary electrophoresis (FACE) (O’Shea, Samuel, Konik, & Morell, 1998) using Proteome Lab
148
PA800 (Beckman Coulter, Fullerton CA, USA) equipped with a 488 mm laser module. A
149
modified debranching protocol (Asare et al., 2011) was used to obtain unit amylopectin chains,
150
which were labeled using 8-aminopyrene 1, 2, 6-trisulfonate in the presence of sodium
151
cyanoborohydride/ tetrahydrofuran. The N-CHO (PVA) capillary with pre-burned window (50
152
μm ID, 50.2 cm total length) was used to separate debranched glucan chains.
153
2.7 Statistical Analysis
154
All biochemical determinations are expressed as average of three biological replicates. The
155
analysis of variance (ANOVA) of the means was performed with SPSS univariate analysis
156
(SPSS Inc. version 17). Multiple means comparisons were determined with the Duncan’s
157
multiple range test at p < 0.05 confidence level. For scanning electron microscopy at least nine
158
areas (from three independent starch preparations) selected on random basis were analyzed and
159
image selected average distribution observed for these preparations.
Ac ce p
te
d
M
an
us
cr
ip t
145
160
161
3. Results and Discussion
162
Wheat endosperm consists of four cell types: the embryo surrounding cells, transfer cells, the
163
aleurone layer and the starchy endosperm (Olsen, 2001). Starchy endosperm in wheat represents
164
the largest body of cell mass in the endosperm consisting of an estimated 60,000 cells (Chojecki,
165
Bayliss, & Gale, 1986). Development of endosperm is broadly divided into five phases 8
Page 8 of 33
(Simmonds & O’Brien, 1981). Phase I (0 DPA) is the fertilization phase. Phase II (1-5 DPA) is
167
the grain growth stage, where the initial endosperm nucleus divides repeatedly forming
168
coenocytic endosperm. Phase III (6-13 DPA) is the initial grain filling stage, when cellularization
169
occurs via formation of radial microtubule systems and alveolation (Olsen, 2004). Phase IV (14-
170
24 DPA) is the maximum grain filling stage, in which rapid starch and protein synthesis occurs
171
in the endosperm (Bechtel, Zayas, Kaleikau, & Pomeranz, 1990). Phase V (25-38 DPA) is the
172
drying stage, leading to desiccation. Morphological and biochemical changes occurring in
173
starchy endosperm during wheat grain development will provide an insight into the factors
174
affecting the final wheat grain quality.
175
3.1 Starch granule morphology during wheat grain development
176
In wheat endosperm, starch granules are produced individually in separate amyloplasts. Starch
177
granules of different shapes and sizes are developed in the endosperm during different periods of
178
grain maturation, and have been divided into different classes. The number of size classes,
179
however, is debated. All reports agree that in the initial grain development period, the first
180
population of starch granules (large A-type, >15 μm) emerges at 4-7 DPA (Bechtel & Wilson,
181
2003). These are initially spherical granules of diameter 0.5-1.0 μm, which grow radially to 2-4
182
μm. According to Evers (1971), an equatorial plate which encircles the granule is formed after
183
the starch granule has reached a specific size. Active deposition of glucan molecules takes place
184
over the equatorial plate, and gives the granule its characteristic lenticular shape. A second
185
population of small starch granules (B-type, 5-15 μm) is initiated at 12-14 DPA, which develop
186
inside amyloplasts protrusions and remain spherical throughout development (Bechtel & Wilson,
187
2003). Some reports suggest the occurrence of a third population of even smaller spherical starch
188
granules (C-type, < 5 μm) at 21 DPA (Raeker, Gaines, Finney, & Donelson, 1998).
Ac ce p
te
d
M
an
us
cr
ip t
166
9
Page 9 of 33
ip t cr us an M d Ac ce p
te
189
10
Page 10 of 33
ip t cr us an M d Fig. 1 A and B
Ac ce p
191
te
190
11
Page 11 of 33
192
The first population of small spherical starch granules (possibly A-type) was observed at 3 and 6
194
DPA, and it increased in size from 9 DPA onwards (Fig. 1A, B). This indicates starch
195
accumulation at the coenocytic stage, before endosperm cell wall formation (Simmonds &
196
O’Brien, 1981). Expression of GBSSI in wheat at this stage has been previously reported
197
(Laudencia-Chingcuanco et al., 2007; Nadaud et al., 2010), suggesting starch formation as early
198
as 3 DPA. In the non-waxy genotype, starch granules of all sizes ranging from 0.8-19.2 μm
199
diameter were initiated at 3 DPA (Fig. 1A). However in the waxy genotype, only small starch
200
granules with diameter maxima of 5.0 μm were observed at this stage (Fig. 1B). At 6 DPA, the
201
starch granule population was dominated by granules up to 11 μm (non-waxy) and 13 μm (waxy)
202
diameters, in addition to smaller starch granules. A second population of starch granules
203
(possibly B-type) of diameter 1.2 μm (non-waxy, Fig. 1A) and 0.8 μm (waxy, Fig. 1B), was
204
observed at 9 and 12 DPA. At 9 DPA, equatorial grooves (as reported by Evers, 1971) were
205
observed for large growing starch granules (Fig. 1). Between 12 and 15 DPA, starch granule
206
diameter maxima reached 23.3 μm (non-waxy, Fig. 1A) and 26.3 μm (waxy, Fig. 1B), developed
207
a shallow equatorial groove, and assumed lenticular shape. Emergence of a new population of
208
small starch granules (possibly C-type) was observed between 18 and 21 DPA as depicted by
209
increased starch granule density of diameter 1.6 μm (non-waxy, Fig. 1A) and 1.7 μm (waxy, Fig.
210
1B). After this stage, large starch granules did not increase in size and maintained a maximum
211
diameter of 22.5 μm for both non-waxy and waxy genotypes. Although both small and large
212
starch granules were present at 30 DPA, the waxy genotype showed a higher number of small
213
starch granules than the non-waxy genotype.
214
3.2 Carbohydrate accumulation during grain development
Ac ce p
te
d
M
an
us
cr
ip t
193
12
Page 12 of 33
Free glucose concentration in the non-waxy genotype was 3.2% at 3 DPA, which decreased to
216
0.2% at maturity. Linear decreases in free glucose concentration, with minor differences, was
217
recorded for all partial waxy and completely waxy genotypes (data not shown), concurring with
218
observations in Brachypodium (Guillon et al., 2011).
219
Starch and amylose contents showed consistent increases with grain maturity, agreeing with
220
previous reports for wheat (Ganeshan, Drinkwater, Repellin, & Chibbar, 2010), barley
221
(McDonald, Stark, Morrison, & Ellis, 1991), and maize (Li, Blanco, & Jane, 2007); however,
222
differing with potato where apparent amylose did not significantly change with maturity (Liu,
223
Weber, Currie, & Yada, 2003). In the present study, starch accumulation increased from 3 to 30
224
DPA, i.e. from 8.3% to 66.2% for non-waxy, and from 3.5% - 5.4% to 53.2% - 66.2% for partial-
225
waxy genotypes. Total starch concentration in the completely waxy genotype increased up to 21
226
DPA (3.9% to 48.0%), after which it reached saturation (50.3%) at 30 DPA (Fig. 2). Starch
227
concentration increased approximately 10% from 12 to 15 DPA in all genotypes, concurring with
228
a previous transcriptional profiling study in wheat which suggested that the maximum
229
accumulation rate of transcripts encoding storage proteins and starch biosynthetic enzymes, as
230
well as maximum starch accumulation, occurs from 7-14 DPA (Laudencia-Chingcuanco et al.,
231
2007).
232
In the present study, the rate of total starch accumulation depicted by linear regression values, in
233
the completely waxy genotype (r2 = 0.90) was higher than the non-waxy genotype (r2 = 0.86),
234
although the non-waxy genotype showed a significantly higher starch concentration throughout
235
development (Fig. 2). Similarly, a high starch accumulation rate and higher enzyme activities of
236
sucrose synthase, AGPase, SS, GBSS and SBE have been reported for a cultivar with higher
237
starch content (Dai, 2010), supporting the higher total starch concentration in the non-
Ac ce p
te
d
M
an
us
cr
ip t
215
13
Page 13 of 33
ip t cr us an M te
Fig. 2
Ac ce p
239
d
238
14
Page 14 of 33
ip t cr us an te
d
Fig. 3
Ac ce p
241
M
240
15
Page 15 of 33
242
waxy genotype in the present study. From 3 to 6 DPA, starch concentration in Wx-A- genotypes
244
doubled, while a small increase was observed in other genotypes. This suggests that Wx-A could
245
be playing a regulatory role in starch accumulation. At 15 and 21 DPA, starch concentration was
246
higher in single nulls than double nulls, suggesting a significant influence of GBSSI dosage on
247
starch accumulation.
248
The amylose accumulation sigmoid-curve during granule development showed a linear increase
249
from 3 to 6 DPA, remained constant till 9 DPA, and increased again from 12 to 18 DPA reaching
250
a stationary phase after 21 DPA (Fig. 3). It has been shown earlier that mRNA of rice GBSSI
251
gene increased with the development of endosperm (Dian, Jiang, & Wu, 2005). Also, increase in
252
GBSSI gene expression is directly related to amylose accumulation in maize endosperm (Li et
253
al., 2007). In the non-waxy genotype, amylose concentration increased from 7.2% at 3 DPA to
254
30.5% at 30 DPA, being higher than partial and completely waxy genotypes (Fig. 3). This was
255
also depicted by higher GBSS activity in the cultivar with higher starch content, indicating that it
256
had a greater capacity to synthesize amylose and amylopectin than the cultivar with low starch
257
content during mid-late grain filling (Dai, 2010). In the present study, the rate of amylose
258
accumulation among the partial waxy genotypes differed during grain development. From 6 to 9
259
DPA, it was higher in the single nulls (r2 = 1.07) than double nulls (r2 = 0.24); from 9 to 18 DPA,
260
it was similar for all the genotypes (r2 = 1.08 - 1.09). Thereafter single nulls had a higher
261
concentration of amylose than double nulls, reaching 29.2% and 24.6% respectively at 30 DPA.
262
The exception to this trend was the completely waxy genotype, which showed a consistent
263
reduction in the rate of amylose accumulation throughout grain development (7.0% to 0.8%)
264
(Fig. 3). Ideally, it should have shown a negligible concentration of amylose at earlier stages of
Ac ce p
te
d
M
an
us
cr
ip t
243
16
Page 16 of 33
grain development. It is speculated that the detected amylose could be due to GBSSII from the
266
pericarp, which is still photosynthetic at that stage. GBSSII is responsible for amylose synthesis
267
in non-storage tissues (Vrinten & Nakamura, 2000), and since it is encoded by a separate gene,
268
absence of GBSSI would not affect amylose accumulation by GBSSII. At the early stages of
269
grain development pericarp could not be separated from the nascent endosperm, therefore
270
amylose was also observed in the waxy starch genotypes.
271
3.3 Starch granule polypeptides during grain development
272
Denaturing polyacrylamide gel analysis for non-waxy and completely waxy genotypes was
273
performed to determine whether the amount of GBSSI accumulated within starch granules
274
correlated with amylose content. For the non-waxy wheat genotype, reduced accumulation of
275
GBSSI (~60 kDa) was observed at 3 and 6 DPA, i.e. at early grain filling stage, while normal
276
accumulation levels were observed from 9 to 30 DPA, or the active grain filling period (Fig. 4A),
277
as has been previously observed (Nadaud et al., 2010). In the non-waxy genotype, the level of
278
accumulation of SSII (100-115 kDa), SBE (87 kDa) and SSI (77 kDa) was lower until 9 DPA,
279
after which a constant level of accumulation of these polypeptides was observed. Very little
280
accumulation of SBEIc (152 kDa) was observed from 18 to 30 DPA (Fig. 4A). In the waxy
281
genotype, a similar level of accumulation of SSII, SBE and SSI was observed throughout grain
282
development. SBEIc could not be detected in the waxy genotype. As expected for the waxy
283
genotype, GBSSI accumulation was observed at 3 and 6 DPA, but not from 9 to 30 DPA (Fig.
284
4B). Presence of GBSSI at early stages of development was confirmed by immunoblot analysis
285
using polyclonal antibodies raised against wheat GBSSI (Supplementary Fig. 1), which was
286
speculated to be GBSSII.
Ac ce p
te
d
M
an
us
cr
ip t
265
17
Page 17 of 33
ip t cr us an M d te Ac ce p 287 288
Fig. 4
18
Page 18 of 33
M
an
us
cr
ip t
289
290
te
d
Fig. 5
Ac ce p
291
19
Page 19 of 33
Since starch synthesizing enzymes in cereals have been shown to work in multi-protein
293
complexes (Tetlow et al., 2008), immunoblotting was performed for other starch synthesizing
294
enzymes to check whether the absence of GBSSI affects the accumulation of other starch granule
295
proteins. SSI, SSII, SBEI and SBEII showed a similar pattern in non-waxy and waxy genotypes
296
indicating that absence of GBSSI did not affect the accumulation of other starch granule bound
297
proteins (Supplementary Fig. 1).
us
cr
ip t
292
298
3.4 Amylopectin chain length distribution during grain development
an
299
M
300
Most variation in amylopectin chain length distribution was observed from 3 to 12 DPA, after
302
which there were no considerable differences up to 30 DPA (Fig. 5). In all genotypes, glucan
303
chain of DP ≤ 14 decreased, while glucan chains of DP ≥ 19 increased during early stages of
304
development. In the non-waxy genotype, glucan chains of DP 6-8, DP 9-11 and DP 12-14
305
decreased from 6.2%, 30.7% and 38.9% at 3 DPA to 3.7%, 25.2% and 32.9% at 9 DPA
306
respectively. Glucan chains of DP 15-18 did not show significant variation across development
307
stages, being 24.2% at 3 DPA, 27.1% at 12 DPA and 25.9% at 30 DPA. Glucan chains of DP 19-
308
22, DP 23-36 and DP 37-45 increased from 2.6%, 0.9% and 1.1% at 6 DPA to 6.6%, 3.6% and
309
3.1% at 12 DPA respectively (data not shown). Glucan chains of DP ≥ 19 were absent at 3 DPA
310
(Fig. 5). These results were also reflected in the increase in average DP from 3 to 12 DPA with
311
no changes thereafter (Supplementary Table 1), suggesting a continuous change in the length and
312
arrangement of glucan chains up to 12 DPA (Fig. 5). The content of short glucan chains was
313
higher during the early endosperm development stage, mainly during the cellularization stage,
Ac ce p
te
d
301
20
Page 20 of 33
and decreased consistently until the initial grain filling stage. From active grain filling onwards,
315
the amylopectin chain length distribution changed only slightly (Fig. 5, Supplementary Table 1).
316
This suggests that amylopectin attains its specific unit chain length distribution/ structure by the
317
grain filling stage. Similarly, in a previous study of endosperm starch properties during maize
318
grain development, it was reported that average amylopectin branch chain length increased from
319
10 DAP to 14 DAP and then decreased on 30 and 45 DAP (Li et al., 2007). It was suggested that
320
the structure of endosperm starch was not synthesized consistently throughout grain
321
development. It has previously been reported that the relative expression of GBSSI gene
322
increases only after 5 DPA (Ganeshan et al., 2010). However other starch synthases (SSI, SSII)
323
genes show a higher relative expression than GBSSI at earlier DPA. For the non-waxy genotype,
324
therefore, the reduction in the content of glucan short chains (Fig. 5) could be a result of
325
increased GBSSI gene expression. As GBSSI gene expression increases, it actively elongates the
326
short glucan chains; hence their number decreases and the proportion of long glucan chains
327
increases. In addition, 11 enzymes associated with starch synthesis, particularly GBSSII and
328
soluble SS, have been shown to have expressed at the very beginning of wheat grain
329
development (Nadaud et al., 2010), which could be the contributing factor for amylopectin
330
synthesis in the absence of GBSSI, in the completely waxy genotype. Starch biosynthetic
331
enzymes activity assays during endosperm development could provide a better understanding of
332
amylopectin structure formation during grain development, as well as the precise function of
333
GBSSI.
334
Subtle differences were observed in the amylopectin glucan chain length distribution with the
335
reduction of waxy protein dosage during wheat grain development (Fig. 6). Compared to non-
Ac ce p
te
d
M
an
us
cr
ip t
314
21
Page 21 of 33
ip t cr us an M d te Ac ce p 336 337
Fig. 6 22
Page 22 of 33
338
waxy starch, completely waxy starch at 3 DPA showed a higher proportion of short glucan
340
chains of DP 6-8 (Fig. 6), suggesting a relatively earlier synthesis of amylopectin chains in the
341
former as compared to the latter, concurring with a previous report (Cao et al., 2012). At 3 DPA,
342
on DP 6-8 glucan chains, the influence of Wx-B, Wx-D alone or together was the highest,
343
whereas, the influence of Wx-A was the lowest. Similarly at 6 DPA, Wx-B, D affected the
344
glucan short chain fraction more than the other partial waxy starches. This suggests that Wx-B
345
and Wx-D are playing a greater role in amylopectin synthesis during the initial stages of grain
346
development. At 9 DPA, however, Wx-A, B showed a higher proportion of DP 6-15 glucan
347
chains followed by Wx-B and Wx-D. However, Wx-A alone was the least effective in
348
influencing DP 6-8 chains, suggesting that Wx-A only in combination with Wx-B or Wx-D has a
349
stronger effect on the short glucan chains formation. Furthermore, at 12 DPA, Wx-A, B
350
displayed a considerably higher proportion of short chains of DP 6-8 than the other partial waxy
351
and non-waxy wheat starches, suggesting a positive influence of Wx-A on the short chain
352
phenotype during grain filling. At 18 and 24 DPA, Wx-D showed an increase in DP 6-8 glucan
353
chains relative to the non-waxy starch. In addition at 24 and 30 DPA, glucan chains of DP > 20
354
were negatively affected by Wx-A (Fig. 6). This suggests that Wx-B and Wx-D play a stronger
355
role in influencing mid-length glucan chains than Wx-A, during later stages of grain
356
development. The above observations suggest that different waxy isoproteins have different
357
capacities for amylopectin structure formation during wheat grain development; Wx-B and Wx-
358
D being the most effective during early grain development, while Wx-A being active as grain
359
filling starts, followed by Wx-D during grain maturation.
360
4. Conclusions
Ac ce p
te
d
M
an
us
cr
ip t
339
23
Page 23 of 33
Endosperm development is a complex process involving numerous physiological and
362
biochemical transitions. Starch granule morphology continuously changed with each phase of
363
development, resulting in consistent starch and amylose accumulation in non-waxy and partial
364
waxy genotypes. Changes in amylopectin structure were observed until 12 DPA, suggesting the
365
formation of a basic amylopectin skeleton by the grain filling stage. Various waxy isoproteins
366
affected amylopectin structure differently during wheat grain development. In the early stages,
367
Wx-B and Wx-D were the most effective, followed by Wx-A during the mid grain-fill stage and
368
Wx-D during grain maturation. Thus, genome specific GBSSI in wheat considerably affects
369
starch accumulation and glucan chain length distribution during grain development.
an
us
cr
ip t
361
M
370
ACKNOWLEDGEMENTS
372
Canada Research Chairs program, Natural Sciences and Engineering Research Council Canada
373
(Discovery grants program), Canada Foundation for Innovation are acknowledged for financial
374
support. GA is a grateful recipient of the Department of Plant Sciences Devolved Graduate
375
Scholarship.
Ac ce p
te
d
371
24
Page 24 of 33
376
REFERENCES
378
AACC International. Approved Methods of Analysis, 11th Ed. Method 76-13.01. Total starch
379
assay procedure (Megazyme amyloglucosidase/alpha-amylase method). St. Paul, MN: AACC
380
International
381
Abdel-Aal, E.-S.M., Hucl, P., Chibbar, R. N., Han, H. L., & Demeke, T. (2002).
382
Physicochemical and structural characteristics of flours and starches from waxy and nonwaxy
383
wheats. Cereal Chemistry, 79, 458-464.
384
Asare, E. K., Jaiswal, S., Maley, J., Båga, M., Sammynaiken, R., Rossnagel, B. G., & Chibbar,
385
R. N. (2011). Barley grain constituents, starch composition, and structure affect starch in vitro
386
enzymatic hydrolysis. Journal of Agricultural and Food Chemistry, 59, 4743-4754.
387
Bechtel, D., Zayas, I., Kaleikau, L., & Pomeranz, Y. (1990). Size-distribution of wheat-starch
388
granules during endosperm development. Cereal Chemistry, 67, 59-63.
389
Bechtel, D. B., & Wilson, J. D. (2003). Amyloplast formation and starch granule development in
390
hard red winter wheat. Cereal Chemistry, 80, 175-183.
391
Buléon, A., Colonna, P., Planchot, V., & Ball, S. (1998). Starch granules: Structure and
392
biosynthesis. International Journal of Biological Macromolecules, 23, 85-112.
393
Cao, Y-N., Hu, W-G., & Wang, C-S. (2012). Relationship among the key enzymatic activities
394
involved in starch synthesis and amylopectin chain distributions in developing wheat grain.
395
African Journal of Biotechnology, 11, 805-814.
Ac ce p
te
d
M
an
us
cr
ip t
377
25
Page 25 of 33
Chojecki, A. J. S., Bayliss, M. W., & Gale, M. D. (1986). Cell production and DNA
397
accumulation in the wheat endosperm, and their association with grain weight. Annals of Botany,
398
58, 809-817.
399
Dai, Z. (2010). Activities of enzymes involved in starch synthesis in wheat grains differing in
400
starch content. Russian Journal of Plant Physiology, 57, 74-78.
401
Delrue, B., Fontaine, T., Routier, F., Decq, A., Wieruszeski, J-M., Koornhuyse, N., Maddelein,
402
M-L., Fournet, B., & Ball, S. (1992). Waxy Chlamydomonas reinhardtii - monocellular algal
403
mutants defective in amylose biosynthesis and granule-bound starch synthase activity
404
accumulate a structurally modified amylopectin. Journal of Bacteriology, 174, 3612-3620.
405
Demeke, T., Hucl, P., Nair, R., Nakamura, T., & Chibbar, R. N. (1997). Evaluation of Canadian
406
and other wheats for waxy proteins. Cereal Chemistry, 74, 442-444.
407
Demeke, T., Hucl, P., Abdel-Aal, E., Båga, M., & Chibbar, R. N. (1999). Biochemical
408
characterization of the wheat waxy A protein and its effect on starch properties. Cereal
409
Chemistry, 76, 694-698.
410
Denyer, K., Clarke, B., Hylton, C., Tatge, H., & Smith, A. (1996). The elongation of amylose
411
and amylopectin chains in isolated starch granules. Plant Journal, 10, 1135-1143.
412
Dian, W., Jiang, H., & Wu, P. (2005). Evolution and expression analysis of starch synthase III
413
and IV in rice. Journal of Experimental Botany, 56, 623-632.
414
Evers, A. D. (1971). Scanning electron microscopy of wheat starch. III. Granule development in
415
the endosperm. Starch/Stärke, 23, 157-160.
Ac ce p
te
d
M
an
us
cr
ip t
396
26
Page 26 of 33
Ganeshan, S., Drinkwater, J. M., Repellin, A., & Chibbar, R. N. (2010). Selected carbohydrate
417
metabolism genes show coincident expression peaks in grains of in vitro-cultured immature
418
spikes of wheat (Triticum aestivum L.). Journal of Agricultural and Food Chemistry, 58, 4193-
419
4201.
420
Gao, M., & Chibbar, R. N. (2000). Isolation, characterization, and expression analysis of starch
421
synthase IIa cDNA from wheat (Triticum aestivum L.). Genome, 43, 768-775.
422
Gromova, I., & Celis, J. E. (2006). Protein detection in gels by silver staining: A procedure
423
compatible with mass-spectroscopy. In J. E. Celis, N. Carter, T. Hunter, K. Simons, J. V. Small,
424
& D. Shotton (Eds.), Cell Biology: A Laboratory Handbook (pp. 1-8). Amsterdam, Boston:
425
Elsevier Academic Press.
426
Guillon, F., Larŕe, C., Petipas, F., Berger, A., Moussawi, J., Rogniaux, H., Santoni, A., Saulnier,
427
L., Jamme, F., Miquel, M., Lepiniec, L., & Dubreucq, B. (2011). A comprehensive overview of
428
grain development in Brachypodium distachyon variety Bd21. Journal of Experimental Botany,
429
63, 739-755.
430
Hizukuri, S. (1986). Polymodal distribution of the chain lengths of amylopectins, and its
431
significance. Carbohydrate Research, 147, 342-347.
432
Hucl, P., & Chibbar, R. N. (1996). Variation for starch concentration in spring wheat and its
433
repeatability relative to protein concentration. Cereal Chemistry, 73, 756-758.
434
Jane, J., Chen, Y. Y., Lee, L. F., McPherson, A. E., Wong, K. S., Radosavljevic, M. &
435
Kasemsuwan, T. (1999). Effects of amylopectin branch chain length and amylose content on the
436
gelatinization and pasting properties of starch. Cereal Chemistry, 76, 629-637.
Ac ce p
te
d
M
an
us
cr
ip t
416
27
Page 27 of 33
Jobling, S. (2004). Improving starch for food and industrial applications. Current Opinion in
438
Plant Biology, 7, 210-218.
439
Kalinga, D. N., Bertoft, E., Tetlow, I., Liu, Q., Yada, R. Y., & Seetharaman, K. (2014)
440
Formation of amylopectin structure in developing wheat endosperm. Carbohydrate Polymers,
441
112, 325-333.
442
Laudencia-Chingcuanco, D. L., Stamova, B. S., You, F. M., Lazo, G. R., Beckles, D. M., &
443
Anderson, O. D. (2007). Transcriptional profiling of wheat caryopsis development using cDNA
444
microarrays. Plant Molecular Biology, 63, 651-668.
445
Li, C-Y, Feng, C-N., Wang, Y-L., Zhang, R., Gua, W-S., Zhu, X-K., & Peng, Y-X. (2007).
446
Chain length distribution of debranched amylopectin and its relationship with physicochemical
447
properties of starch in different wheat cultivars. Acta Agronomica Sinica, 33, 1240-1245.
448
Li, L., Blanco, M., & Jane, J-L. (2007). Physicochemical properties of endosperm and pericarp
449
starches during maize development. Carbohydrate Polymers, 67, 630-639.
450
Liu, Q., Weber, E., Currie, V., & Yada, R. (2003). Physicochemical properties of starches during
451
potato growth. Carbohydrate Polymers, 51, 213-221.
452
Manners, D. J. (1989). Recent developments in our understanding of amylopectin structure.
453
Carbohydrate Polymers, 11, 87-112.
454
McCleary, B., Gibson, T., & Mugford, D. (1997). Measurement of total starch in cereal products
455
by amyloglucosidase-alpha-amylase method: Collaborative study. Journal of AOAC
456
International, 80, 571-579.
Ac ce p
te
d
M
an
us
cr
ip t
437
28
Page 28 of 33
McDonald, A. M. L., Stark, J. R., Morrison, W. R., & Ellis, R. P. (1991). The composition of
458
starch granules from developing barley genotypes. Journal of Cereal Science, 13, 93-112.
459
Nadaud, I., Girousse, C., Debiton, C., Chambon, C., Bouzidi, M. F., Martre, P., & Branlard, G.
460
(2010). Proteomic and morphological analysis of early stages of wheat grain development.
461
Proteomics, 10, 2901-2910.
462
Olsen, O-A. (2001). Endosperm development: Cellularization and cell fate specification. Annual
463
Review of Plant Physiology and Plant Molecular Biology, 52, 233-267.
464
Olsen, O-A. (2004). Nuclear endosperm development in cereals and Arabidopsis thaliana. Plant
465
Cell, 16, S214-S227.
466
O'Shea, M., Samuel, M., Konik, C., & Morell, M. (1998). Fluorophore-assisted carbohydrate
467
electrophoresis (FACE) of oligosaccharides: Efficiency of labelling and high-resolution
468
separation. Carbohydrate Research, 307, 1-12.
469
Peng, M., Gao, M., Abdel-Aal, E., Hucl, P., & Chibbar, R. N. (1999) Separation and
470
characterization of A- and B-type starch granules in wheat endosperm. Cereal Chemistry, 76,
471
375-379.
472
Peng, M., Hucl, P., & Chibbar, R.N. (2001) Isolation, characterization and expression analysis of
473
starch synthase I from wheat (Triticum aestivum L.). Plant Science 161, 1055–1062.
474
Raeker, M., Gaines, C., Finney, P., & Donelson, T. (1998). Granule size distribution and
475
chemical composition of starches from 12 soft wheat cultivars. Cereal Chemistry, 75, 721-728.
Ac ce p
te
d
M
an
us
cr
ip t
457
29
Page 29 of 33
Ral, J-P., Colleoni, C., Wattebled, F., Dauvillée, D., Nempont, C., Deschamps, P., Li, Z., Morell,
477
M. K., Chibbar, R. N., Purton, S., d’Hulst, C., & Ball, S. (2006). Circadian clock regulation of
478
starch metabolism establishes GBSSI as a major contributor to amylopectin synthesis in
479
Chlamydomonas reinhardtii. Plant Physiology, 142, 305-317.
480
Sharma, R., Sissons, M. J., Rathjen, A. J., & Jenner, C. F. (2002). The null-4A allele at the waxy
481
locus in durum wheat affects pasta cooking quality. Journal of Cereal Science, 35, 287-297.
482
Simmonds, D., & O’Brien, T. (1981). Morphological and biochemical development of the wheat
483
endosperm. In Y. Pomeranz (Ed.), Advances in Cereal Science and Technology Vol. 4 (pp. 5 –
484
70). St. Paul, MN: American Association of Cereal Chemists.
485
Tetlow, I. J., Beisel, K. G., Cameron, S., Makhmoudova, A., Liu, F., Bresolin, N. S., Wait, R.,
486
Morell, M. K., & Emes, M. J. (2008). Analysis of protein complexes in wheat amyloplasts
487
reveals functional interactions among starch biosynthetic enzymes. Plant Physiology, 146, 1878-
488
1891.
489
Vrinten, P. L., & Nakamura, T. (2000). Wheat granule-bound starch synthase I and II are
490
encoded by separate genes that are expressed in different tissues. Plant Physiology, 122, 255-
491
263.
492
Waduge, R. N., Xu, S., Bertoft, E., & Seetharaman, K. (2013) Exploring the surface morphology
493
of developing wheat starch granules by using Atomic Force Microscopy. Starch/ Stärke, 65, 398-
494
409.
Ac ce p
te
d
M
an
us
cr
ip t
476
30
Page 30 of 33
Yasui, T., Matsuki, J., Sasaki, T., & Yamamori, M. (1996). Amylose and lipid contents,
496
amylopectin structure, and gelatinisation properties of waxy wheat (Triticum aestivum) starch.
497
Journal of Cereal Science, 24, 131-137.
498
Zhao, X. C., Batey, I. L., Sharp, P. J., Crosbie, G., Barclay, I., Wilson, R., Morell, M. K., &
499
Appels, R. (1998). A single genetic locus associated with starch granule properties and noodle
500
quality in wheat. Journal of Cereal Science, 27, 7-13.
Ac ce p
te
d
M
an
us
cr
ip t
495
31
Page 31 of 33
501
FIGURE CAPTIONS
503
Fig. 1 SEM images of starch granules isolated at different development stages in non-waxy (A)
504
and completely waxy (B) genotypes.
505
Fig. 2 Total starch concentration in near isogenic (waxy locus) lines during wheat grain
506
development.
507
Fig. 3 Amylose concentration in near isogenic (waxy locus) lines during wheat grain
508
development.
509
Fig. 4 Denaturing gel electrophoresis analysis showing the presence of starch granule-bound
510
proteins (silver staining) during grain development (3 to 30 DPA) in non-waxy (A) and
511
completely waxy (B) genotypes.
512
Fig. 5 Amylopectin chain length distribution during wheat grain development (3 – 30 DPA) in a
513
non-waxy genotype.
514
Fig. 6 Amylopectin chain length distribution in partial and completely waxy wheat genotypes
515
during grain development (3, 6, 9, 12, 18, 24, and 30 DPA). Subtractive graphs depict relative
516
area in a particular waxy-null genotype subtracted from that of normal genotype. Values are
517
based on average of three biological replicates.
Ac ce p
te
d
M
an
us
cr
ip t
502
32
Page 32 of 33
519
520
521
Highlights
ip t
522
1. Wheat GBSSI ( (Wx) dosage influence starch accumulation during grain development
524
2. Wx-A influences starch accumulation at early stages of grain development
525
3. Wx-B and Wx-D affects synthesis of amylopectin mid-length glucan chains
526
4. Waxy starch amylopectin has higher fibril height than non-waxy
us
cr
523
Ac ce p
te
d
M
an
527
34
Page 33 of 33
S0144-8617(14)00761-9 http://dx.doi.org/doi:10.1016/j.carbpol.2014.08.004 CARP 9131
To appear in: Received date: Revised date: Accepted date:
28-3-2014 1-8-2014 4-8-2014
Please cite this article as: Ahuja, G.,Wheat genome specific granule-bound starch synthase I differentially influence grain starch synthesis, Carbohydrate Polymers (2014), http://dx.doi.org/10.1016/j.carbpol.2014.08.004 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Wheat genome specific granule-bound starch synthase I differentially influence grain
2
starch synthesis
3
Authors: Geetika Ahuja, Sarita Jaiswal, Pierre Hucl, Ravindra N. Chibbar
4
Affiliation address: Department of Plant Sciences & Crop Development Centre, College of
5
Agriculture and Bioresources, 51 Campus Drive, University of Saskatchewan, Saskatoon SK
6
S7N 5A8, Canada
7
Email addresses: [email protected], [email protected], [email protected],
8
[email protected]
9
Corresponding author: Professor Ravindra N. Chibbar, Department of Plant Sciences & Crop
10
Development Centre, College of Agriculture and Bioresources, 51 Campus Drive, University of
11
Saskatchewan, Saskatoon SK S7N 5A8, Canada
12
E-mail: [email protected]
13
Phone: +1-306-966-4969
14
Fax: +1-306 966-5015
Ac ce p
te
d
M
an
us
cr
ip t
1
1
Page 1 of 33
Abstract
16
Wheat grain development is a complex process and is characterized by changes in
17
physicochemical and structural properties of starch. The present study deals with endosperm
18
starch physicochemical properties and structure during development in different granule-bound
19
starch synthase I (GBSSI) null also known as waxy (Wx) genotypes. The study was conducted
20
with pure starch isolated from wheat grains at 3-30 days post anthesis (DPA), at three day
21
intervals. Amylose concentration increased throughout grain development in non-waxy (7.2% –
22
30.5%) and partial waxy genotypes (6.0%– 26.8%). Completely waxy genotype showed 7.0%
23
amylose at 3 and 6 DPA, which declined during development and reached non-detectable
24
quantities by 30 DPA. Amylopectin structure had a higher content of short chains at 3 DPA,
25
which decreased continuously until 12 DPA, after which there were only minor changes in
26
amylopectin chain length distribution. Similarly, the average degree of polymerization (DP)
27
increased from 3 DPA (12.3) to 12 DPA (15.0), and then did not differ significantly up to 30
28
DPA (15.0). This suggests the formation of basic amylopectin architecture in wheat by 12 DPA.
29
Wx-B and Wx-D affected amylopectin short chains mostly of DP 6-8 at 3 and 6 DPA. Wx-A
30
affected the same fraction of chains at 9 and 12 DPA, and Wx-D affected DP 18-25 chains from
31
18 to 30 DPA, suggesting differential effect of waxy isoproteins on amylopectin structure
32
formation.
33
Keywords: Granule-bound starch synthase I, waxy, grain development, amylopectin chain
34
length distribution.
35
Abbreviations1
Ac ce p
te
d
M
an
us
cr
ip t
15
1
DPA – Days post anthesis, DP – Degree of polymerization, GBSSI – Granule-bound starch synthase I, SEM –
Scanning electron microscopy
2
Page 2 of 33
36
1. Introduction
38
Biochemical properties of starch have been extensively utilized in the food industry for uses such
39
as thickeners for soups, sauces and dairy products, starch-based fat replacers to maintain texture,
40
flavor and mouth-feel of low- or no-fat products. The relative proportion and structure of starch
41
constituents, amylose and amylopectin, greatly influence its end-use. For instance, waxy wheat
42
starches possess exceptional freeze-thaw stability, water absorption capacity, peak viscosity, and
43
low gelatinization temperatures; hence they are useful in the baking industry (Abdel-Aal, Hucl,
44
Chibbar, Han, & Demeke, 2002). Amylopectin chain length distribution also affects
45
gelatinization, retrogradation, and pasting properties of starch. Short glucan chains reduce
46
gelatinization temperatures (Jobling, 2004), while very-long branch chains influence starch-
47
pasting properties (Jane et al., 1999); thus making them valuable in food industry applications.
48
Structurally, starch is a glucan homopolymer composed of one-quarter amylose (linear with few
49
branches) and three-quarters amylopectin (highly branched). Linear chains are formed by α-1,4
50
glucosidic linkages, while branches are formed by α-1,6 glucosidic linkages. The structures of
51
amylose and amylopectin, and their organization in the intact granule have been widely
52
discussed (Buléon, Colonna, Planchot, & Ball, 1998). Amylopectin shows a polymodal
53
distribution of chains, organized into clusters, giving the molecule its architectural specificity
54
(Hizukuri, 1986). Briefly, A-chains (DP 6-12) are the outermost chains, B-chains named B1, B2
55
and B3 depending on their degree of polymerization (DP 13-24 up to 50) support A-chains or
56
other B-chains. They are attached via α-1,6 glucosidic linkage, to the central C-chain, which
57
within a granule is oriented towards the centre or hilum of the granule (Manners, 1989). Packing
Ac ce p
te
d
M
an
us
cr
ip t
37
3
Page 3 of 33
of these chains, or rather the amylopectin molecule in the granule determines the basic structure
59
of the starch granule.
60
The first step in the synthesis of amylose and amylopectin is brought about by ADP-glucose
61
pyrophosphorylase (AGPase). Elongation and branching of amylopectin is a complex process
62
and it requires an array of enzymes viz., starch synthases (SS), starch branching enzymes (SBE)
63
and debranching enzymes (DBE) (Buléon et al., 1998). Synthesis of amylose, however, is
64
brought about solely by the enzyme granule-bound starch synthase I (GBSSI) or waxy protein.
65
Plants lacking GBSSI are termed ‘waxy’. Recently, GBSSI has also been postulated to have a
66
role in amylopectin synthesis (Ral et al., 2006). Incorporation of [14C] glucose from ADP [14C]
67
glucose by GBSSI in granules from different species has been reported to be mainly into the long
68
chain fraction of amylopectin (Denyer, Clarke, Hylton, Tatge, & Smith, 1996). However, the
69
precise contribution of GBSSI to amylopectin synthesis is still unclear.
70
Contrasting reports exist on the differences in amylopectin chain length distribution due to the
71
absence of waxy protein. In Chlamydomonas and wheat waxy mutants that lack GBSSI, the
72
amylopectin molecule has been shown to differ from that of their non-waxy counterparts (Delrue
73
et al., 1992; Li et al., 2007). In contrast, no obvious difference in the glucan chain-length
74
distribution profiles of amylopectin molecules between waxy and non-waxy starch within three
75
chain-length groups were reported (Yasui, Matsuki, Sasaki, & Yamamori, 1996). These studies
76
were conducted on mature grains. To better understand changes in amylopectin chain length
77
distribution profiles, some studies have been conducted at different developmental stages.
78
(Waduge, Xu, Bertoft & Seetharaman, 2013; Kalinga, Tetlow, Liu, Yada & Seetharaman, 2014).
79
It was reported that the content of A- and short B-glucan chains increased in the waxy, and
80
decreased in the non-waxy type of wheat during grain development. Also, the content of mid-
Ac ce p
te
d
M
an
us
cr
ip t
58
4
Page 4 of 33
length B-glucan chains decreased in both waxy and non-waxy starch during grain development.
82
It was also suggested that GBSSI elongates the A- or short B-glucan chains to produce long B-
83
glucan chains of amylopectin (Cao, Hu, & Wang, 2012).
84
Since wheat is an allohexaploid (2n = 6x = 42), it has three doses of GBSSI, one from each
85
genome, namely Wx-A, Wx-B and Wx-D. The lack of waxy protein from each genome has been
86
shown to affect the end-use of starch. Durum wheat, which possesses waxy protein from two
87
genomes, is the best candidate for pasta production (Sharma, Sissons, Rathjen, & Jenner, 2002).
88
Similarly, double null-waxy mutants of wheat result in better texture of noodles (Zhao et al.,
89
1998). Considering GBSSI potentially contributes to amylopectin synthesis or elongation, its
90
absence from each genome could affect the glucan chain length distribution of amylopectin
91
and/or organization of glucan chains within the granule, resulting in changes in the morphology
92
of starch granules during wheat grain development. Therefore, the present study focused on
93
studying starch characteristics during wheat grain development in non-waxy and waxy-null
94
genotypes.
95
2. Materials and Methods
96
2.1 Plant material
97
Near-isogenic partial-waxy and non-waxy genotypes including four lines each of Wx-A+B+D+,
98
A-B+D+, A+B-D+, A+B+D-, A-B-D+, A-B+D-, A+B-D- and A-B-D- were used for the developmental
99
study. Plants were grown in a growth chamber in pots containing Redi-earth (W. R. Grace & Co.
Ac ce p
te
d
M
an
us
cr
ip t
81
100
of Canada, ON, Canada) at 23°C / 16 h light (350 μmol m-2 s-2 PPFD) and 19 °C / 8 h dark, and
101
were fertilized every 3 weeks with a slow release fertilizer, Nutricote-14-14-14:N-P-K (Plant
102
Products Co. Ltd., Brampton ON, Canada). Spikes, in triplicate, were tagged at anthesis, 5
Page 5 of 33
collected and frozen in liquid N2 at selected days post anthesis (DPA), viz., 3, 6, 9, 12, 15, 18,
104
21, 24, 27 and 30 days, and stored at -80°C for further experiments. Spikes were collected in
105
mid-afternoon consistently for all developmental stages. Starch was purified from freeze-dried
106
seeds from the middle of the spikes using a modified method (Peng, Gao, Abdel-Aal, Hucl, &
107
Chibbar, 1999) involving cesium chloride density gradient centrifugation (Asare et al., 2011).
108
2.2 Starch granule morphology by scanning electron microscopy
109
Prior to imaging under scanning electron microscope (SEM), starch granules were gold coated to
110
make them electrically conductive. Starch granules were spread on, or attached to a metal stub
111
with double-sided carbon tape and gold coated using a sputter gold coater (S150B, Edwards,
112
Wilmington MA, USA). Granule morphology was observed using a scanning electron
113
microscope (SU6600 Hitachi High-Technologies Europe GmbH, Krefeld, Germany) at 5 kV.
114
2.3 Total starch concentration determination
115
Total starch concentration was determined using an American Association of Cereal Chemists
116
International (AACCI) approved method (AACC method 76-13.01), in which ground meal
117
samples were hydrolyzed to dextrins and finally to D-glucose using α-amylase and
118
amyloglucosidase respectively. Total starch concentration was determined as free glucose by
119
measuring the absorbance at 510 nm (Hucl & Chibbar, 1996) and calculated on a percent dry
120
weight basis (McCleary, Gibson, & Mugford, 1997).
121
2.4 Amylose concentration determination by HPSEC
122
High performance size exclusion chromatography (HPSEC) was used to determine the amylose
123
concentration (Demeke, Hucl, Abdel-Aal, Båga, & Chibbar, 1999). Gelatinized starch samples
Ac ce p
te
d
M
an
us
cr
ip t
103
6
Page 6 of 33
(1mg/mL) were debranched using isoamylase one unit for each mg of starch (1000U/mL
125
Megazyme International Ltd., Wicklow, Ireland), freeze-dried, suspended in dimethyl sulfoxide
126
(DMSO) and injected (40 μL) into the column (PL gel MiniMix 250 x 4.6 mm ID column,
127
Polymer Laboratories Inc., Amherst MA, USA) attached with Waters 410 differential
128
refractometer . (Asare et al., 2011). Starch samples, column, and detector were maintained at 40,
129
100, and 45°C, respectively. DMSO (99%) with 0.4 % Lithium bromide was used as an eluent at
130
a flow rate of 0.2 mL/min. Data was collected and analyzed using Empower 1154
131
chromatography software (Waters Corporation, Milford MA, USA).
132
2.5 Protein detection and immunoblotting
133
Proteins were extracted from the starch granules by suspending 10 mg starch in 300 μL
134
denaturing electrophoresis buffer (Demeke, Hucl, Nair, Nakamura, & Chibbar, 1997). GBSSI
135
was separated using polyacrylamide {15% (w/v) 1 mm thick gel} denaturing gel electrophoresis
136
(Protean II, Biorad, Hercules CA, USA) (Demeke, Hucl, Nair, Nakamura, & Chibbar, 1997).
137
Separated polypeptides were visualised by silver staining (Gromova & Celis, 2006). The gel-
138
electrophoresis separated polypeptides were transferred on a nitrocellulose membrane at 10 V for
139
4 h (Gao & Chibbar, 2000). The membrane with transferred polypeptides was processed and
140
incubated with primary antibodies raised against wheat GBSSI, SSI, SSII and SBE (Gao &
141
Chibbar, 2000; Peng, Hucl, & Chibbar, 2001). The antigen-antibody complex was then detected
142
with goat anti-rabbit alkaline phosphate conjugate using a 5-bromo-4-chloro3’-indolyphosphate
143
p-toluidine / nitro-blue tetrazolium chloride immunoscreening color development kit (Biorad,
144
Hercules CA, USA).
Ac ce p
te
d
M
an
us
cr
ip t
124
7
Page 7 of 33
2.6 Amylopectin chain length distribution by capillary electrophoresis
146
Chain length distribution of amylopectin molecules was determined by fluorophore assisted
147
capillary electrophoresis (FACE) (O’Shea, Samuel, Konik, & Morell, 1998) using Proteome Lab
148
PA800 (Beckman Coulter, Fullerton CA, USA) equipped with a 488 mm laser module. A
149
modified debranching protocol (Asare et al., 2011) was used to obtain unit amylopectin chains,
150
which were labeled using 8-aminopyrene 1, 2, 6-trisulfonate in the presence of sodium
151
cyanoborohydride/ tetrahydrofuran. The N-CHO (PVA) capillary with pre-burned window (50
152
μm ID, 50.2 cm total length) was used to separate debranched glucan chains.
153
2.7 Statistical Analysis
154
All biochemical determinations are expressed as average of three biological replicates. The
155
analysis of variance (ANOVA) of the means was performed with SPSS univariate analysis
156
(SPSS Inc. version 17). Multiple means comparisons were determined with the Duncan’s
157
multiple range test at p < 0.05 confidence level. For scanning electron microscopy at least nine
158
areas (from three independent starch preparations) selected on random basis were analyzed and
159
image selected average distribution observed for these preparations.
Ac ce p
te
d
M
an
us
cr
ip t
145
160
161
3. Results and Discussion
162
Wheat endosperm consists of four cell types: the embryo surrounding cells, transfer cells, the
163
aleurone layer and the starchy endosperm (Olsen, 2001). Starchy endosperm in wheat represents
164
the largest body of cell mass in the endosperm consisting of an estimated 60,000 cells (Chojecki,
165
Bayliss, & Gale, 1986). Development of endosperm is broadly divided into five phases 8
Page 8 of 33
(Simmonds & O’Brien, 1981). Phase I (0 DPA) is the fertilization phase. Phase II (1-5 DPA) is
167
the grain growth stage, where the initial endosperm nucleus divides repeatedly forming
168
coenocytic endosperm. Phase III (6-13 DPA) is the initial grain filling stage, when cellularization
169
occurs via formation of radial microtubule systems and alveolation (Olsen, 2004). Phase IV (14-
170
24 DPA) is the maximum grain filling stage, in which rapid starch and protein synthesis occurs
171
in the endosperm (Bechtel, Zayas, Kaleikau, & Pomeranz, 1990). Phase V (25-38 DPA) is the
172
drying stage, leading to desiccation. Morphological and biochemical changes occurring in
173
starchy endosperm during wheat grain development will provide an insight into the factors
174
affecting the final wheat grain quality.
175
3.1 Starch granule morphology during wheat grain development
176
In wheat endosperm, starch granules are produced individually in separate amyloplasts. Starch
177
granules of different shapes and sizes are developed in the endosperm during different periods of
178
grain maturation, and have been divided into different classes. The number of size classes,
179
however, is debated. All reports agree that in the initial grain development period, the first
180
population of starch granules (large A-type, >15 μm) emerges at 4-7 DPA (Bechtel & Wilson,
181
2003). These are initially spherical granules of diameter 0.5-1.0 μm, which grow radially to 2-4
182
μm. According to Evers (1971), an equatorial plate which encircles the granule is formed after
183
the starch granule has reached a specific size. Active deposition of glucan molecules takes place
184
over the equatorial plate, and gives the granule its characteristic lenticular shape. A second
185
population of small starch granules (B-type, 5-15 μm) is initiated at 12-14 DPA, which develop
186
inside amyloplasts protrusions and remain spherical throughout development (Bechtel & Wilson,
187
2003). Some reports suggest the occurrence of a third population of even smaller spherical starch
188
granules (C-type, < 5 μm) at 21 DPA (Raeker, Gaines, Finney, & Donelson, 1998).
Ac ce p
te
d
M
an
us
cr
ip t
166
9
Page 9 of 33
ip t cr us an M d Ac ce p
te
189
10
Page 10 of 33
ip t cr us an M d Fig. 1 A and B
Ac ce p
191
te
190
11
Page 11 of 33
192
The first population of small spherical starch granules (possibly A-type) was observed at 3 and 6
194
DPA, and it increased in size from 9 DPA onwards (Fig. 1A, B). This indicates starch
195
accumulation at the coenocytic stage, before endosperm cell wall formation (Simmonds &
196
O’Brien, 1981). Expression of GBSSI in wheat at this stage has been previously reported
197
(Laudencia-Chingcuanco et al., 2007; Nadaud et al., 2010), suggesting starch formation as early
198
as 3 DPA. In the non-waxy genotype, starch granules of all sizes ranging from 0.8-19.2 μm
199
diameter were initiated at 3 DPA (Fig. 1A). However in the waxy genotype, only small starch
200
granules with diameter maxima of 5.0 μm were observed at this stage (Fig. 1B). At 6 DPA, the
201
starch granule population was dominated by granules up to 11 μm (non-waxy) and 13 μm (waxy)
202
diameters, in addition to smaller starch granules. A second population of starch granules
203
(possibly B-type) of diameter 1.2 μm (non-waxy, Fig. 1A) and 0.8 μm (waxy, Fig. 1B), was
204
observed at 9 and 12 DPA. At 9 DPA, equatorial grooves (as reported by Evers, 1971) were
205
observed for large growing starch granules (Fig. 1). Between 12 and 15 DPA, starch granule
206
diameter maxima reached 23.3 μm (non-waxy, Fig. 1A) and 26.3 μm (waxy, Fig. 1B), developed
207
a shallow equatorial groove, and assumed lenticular shape. Emergence of a new population of
208
small starch granules (possibly C-type) was observed between 18 and 21 DPA as depicted by
209
increased starch granule density of diameter 1.6 μm (non-waxy, Fig. 1A) and 1.7 μm (waxy, Fig.
210
1B). After this stage, large starch granules did not increase in size and maintained a maximum
211
diameter of 22.5 μm for both non-waxy and waxy genotypes. Although both small and large
212
starch granules were present at 30 DPA, the waxy genotype showed a higher number of small
213
starch granules than the non-waxy genotype.
214
3.2 Carbohydrate accumulation during grain development
Ac ce p
te
d
M
an
us
cr
ip t
193
12
Page 12 of 33
Free glucose concentration in the non-waxy genotype was 3.2% at 3 DPA, which decreased to
216
0.2% at maturity. Linear decreases in free glucose concentration, with minor differences, was
217
recorded for all partial waxy and completely waxy genotypes (data not shown), concurring with
218
observations in Brachypodium (Guillon et al., 2011).
219
Starch and amylose contents showed consistent increases with grain maturity, agreeing with
220
previous reports for wheat (Ganeshan, Drinkwater, Repellin, & Chibbar, 2010), barley
221
(McDonald, Stark, Morrison, & Ellis, 1991), and maize (Li, Blanco, & Jane, 2007); however,
222
differing with potato where apparent amylose did not significantly change with maturity (Liu,
223
Weber, Currie, & Yada, 2003). In the present study, starch accumulation increased from 3 to 30
224
DPA, i.e. from 8.3% to 66.2% for non-waxy, and from 3.5% - 5.4% to 53.2% - 66.2% for partial-
225
waxy genotypes. Total starch concentration in the completely waxy genotype increased up to 21
226
DPA (3.9% to 48.0%), after which it reached saturation (50.3%) at 30 DPA (Fig. 2). Starch
227
concentration increased approximately 10% from 12 to 15 DPA in all genotypes, concurring with
228
a previous transcriptional profiling study in wheat which suggested that the maximum
229
accumulation rate of transcripts encoding storage proteins and starch biosynthetic enzymes, as
230
well as maximum starch accumulation, occurs from 7-14 DPA (Laudencia-Chingcuanco et al.,
231
2007).
232
In the present study, the rate of total starch accumulation depicted by linear regression values, in
233
the completely waxy genotype (r2 = 0.90) was higher than the non-waxy genotype (r2 = 0.86),
234
although the non-waxy genotype showed a significantly higher starch concentration throughout
235
development (Fig. 2). Similarly, a high starch accumulation rate and higher enzyme activities of
236
sucrose synthase, AGPase, SS, GBSS and SBE have been reported for a cultivar with higher
237
starch content (Dai, 2010), supporting the higher total starch concentration in the non-
Ac ce p
te
d
M
an
us
cr
ip t
215
13
Page 13 of 33
ip t cr us an M te
Fig. 2
Ac ce p
239
d
238
14
Page 14 of 33
ip t cr us an te
d
Fig. 3
Ac ce p
241
M
240
15
Page 15 of 33
242
waxy genotype in the present study. From 3 to 6 DPA, starch concentration in Wx-A- genotypes
244
doubled, while a small increase was observed in other genotypes. This suggests that Wx-A could
245
be playing a regulatory role in starch accumulation. At 15 and 21 DPA, starch concentration was
246
higher in single nulls than double nulls, suggesting a significant influence of GBSSI dosage on
247
starch accumulation.
248
The amylose accumulation sigmoid-curve during granule development showed a linear increase
249
from 3 to 6 DPA, remained constant till 9 DPA, and increased again from 12 to 18 DPA reaching
250
a stationary phase after 21 DPA (Fig. 3). It has been shown earlier that mRNA of rice GBSSI
251
gene increased with the development of endosperm (Dian, Jiang, & Wu, 2005). Also, increase in
252
GBSSI gene expression is directly related to amylose accumulation in maize endosperm (Li et
253
al., 2007). In the non-waxy genotype, amylose concentration increased from 7.2% at 3 DPA to
254
30.5% at 30 DPA, being higher than partial and completely waxy genotypes (Fig. 3). This was
255
also depicted by higher GBSS activity in the cultivar with higher starch content, indicating that it
256
had a greater capacity to synthesize amylose and amylopectin than the cultivar with low starch
257
content during mid-late grain filling (Dai, 2010). In the present study, the rate of amylose
258
accumulation among the partial waxy genotypes differed during grain development. From 6 to 9
259
DPA, it was higher in the single nulls (r2 = 1.07) than double nulls (r2 = 0.24); from 9 to 18 DPA,
260
it was similar for all the genotypes (r2 = 1.08 - 1.09). Thereafter single nulls had a higher
261
concentration of amylose than double nulls, reaching 29.2% and 24.6% respectively at 30 DPA.
262
The exception to this trend was the completely waxy genotype, which showed a consistent
263
reduction in the rate of amylose accumulation throughout grain development (7.0% to 0.8%)
264
(Fig. 3). Ideally, it should have shown a negligible concentration of amylose at earlier stages of
Ac ce p
te
d
M
an
us
cr
ip t
243
16
Page 16 of 33
grain development. It is speculated that the detected amylose could be due to GBSSII from the
266
pericarp, which is still photosynthetic at that stage. GBSSII is responsible for amylose synthesis
267
in non-storage tissues (Vrinten & Nakamura, 2000), and since it is encoded by a separate gene,
268
absence of GBSSI would not affect amylose accumulation by GBSSII. At the early stages of
269
grain development pericarp could not be separated from the nascent endosperm, therefore
270
amylose was also observed in the waxy starch genotypes.
271
3.3 Starch granule polypeptides during grain development
272
Denaturing polyacrylamide gel analysis for non-waxy and completely waxy genotypes was
273
performed to determine whether the amount of GBSSI accumulated within starch granules
274
correlated with amylose content. For the non-waxy wheat genotype, reduced accumulation of
275
GBSSI (~60 kDa) was observed at 3 and 6 DPA, i.e. at early grain filling stage, while normal
276
accumulation levels were observed from 9 to 30 DPA, or the active grain filling period (Fig. 4A),
277
as has been previously observed (Nadaud et al., 2010). In the non-waxy genotype, the level of
278
accumulation of SSII (100-115 kDa), SBE (87 kDa) and SSI (77 kDa) was lower until 9 DPA,
279
after which a constant level of accumulation of these polypeptides was observed. Very little
280
accumulation of SBEIc (152 kDa) was observed from 18 to 30 DPA (Fig. 4A). In the waxy
281
genotype, a similar level of accumulation of SSII, SBE and SSI was observed throughout grain
282
development. SBEIc could not be detected in the waxy genotype. As expected for the waxy
283
genotype, GBSSI accumulation was observed at 3 and 6 DPA, but not from 9 to 30 DPA (Fig.
284
4B). Presence of GBSSI at early stages of development was confirmed by immunoblot analysis
285
using polyclonal antibodies raised against wheat GBSSI (Supplementary Fig. 1), which was
286
speculated to be GBSSII.
Ac ce p
te
d
M
an
us
cr
ip t
265
17
Page 17 of 33
ip t cr us an M d te Ac ce p 287 288
Fig. 4
18
Page 18 of 33
M
an
us
cr
ip t
289
290
te
d
Fig. 5
Ac ce p
291
19
Page 19 of 33
Since starch synthesizing enzymes in cereals have been shown to work in multi-protein
293
complexes (Tetlow et al., 2008), immunoblotting was performed for other starch synthesizing
294
enzymes to check whether the absence of GBSSI affects the accumulation of other starch granule
295
proteins. SSI, SSII, SBEI and SBEII showed a similar pattern in non-waxy and waxy genotypes
296
indicating that absence of GBSSI did not affect the accumulation of other starch granule bound
297
proteins (Supplementary Fig. 1).
us
cr
ip t
292
298
3.4 Amylopectin chain length distribution during grain development
an
299
M
300
Most variation in amylopectin chain length distribution was observed from 3 to 12 DPA, after
302
which there were no considerable differences up to 30 DPA (Fig. 5). In all genotypes, glucan
303
chain of DP ≤ 14 decreased, while glucan chains of DP ≥ 19 increased during early stages of
304
development. In the non-waxy genotype, glucan chains of DP 6-8, DP 9-11 and DP 12-14
305
decreased from 6.2%, 30.7% and 38.9% at 3 DPA to 3.7%, 25.2% and 32.9% at 9 DPA
306
respectively. Glucan chains of DP 15-18 did not show significant variation across development
307
stages, being 24.2% at 3 DPA, 27.1% at 12 DPA and 25.9% at 30 DPA. Glucan chains of DP 19-
308
22, DP 23-36 and DP 37-45 increased from 2.6%, 0.9% and 1.1% at 6 DPA to 6.6%, 3.6% and
309
3.1% at 12 DPA respectively (data not shown). Glucan chains of DP ≥ 19 were absent at 3 DPA
310
(Fig. 5). These results were also reflected in the increase in average DP from 3 to 12 DPA with
311
no changes thereafter (Supplementary Table 1), suggesting a continuous change in the length and
312
arrangement of glucan chains up to 12 DPA (Fig. 5). The content of short glucan chains was
313
higher during the early endosperm development stage, mainly during the cellularization stage,
Ac ce p
te
d
301
20
Page 20 of 33
and decreased consistently until the initial grain filling stage. From active grain filling onwards,
315
the amylopectin chain length distribution changed only slightly (Fig. 5, Supplementary Table 1).
316
This suggests that amylopectin attains its specific unit chain length distribution/ structure by the
317
grain filling stage. Similarly, in a previous study of endosperm starch properties during maize
318
grain development, it was reported that average amylopectin branch chain length increased from
319
10 DAP to 14 DAP and then decreased on 30 and 45 DAP (Li et al., 2007). It was suggested that
320
the structure of endosperm starch was not synthesized consistently throughout grain
321
development. It has previously been reported that the relative expression of GBSSI gene
322
increases only after 5 DPA (Ganeshan et al., 2010). However other starch synthases (SSI, SSII)
323
genes show a higher relative expression than GBSSI at earlier DPA. For the non-waxy genotype,
324
therefore, the reduction in the content of glucan short chains (Fig. 5) could be a result of
325
increased GBSSI gene expression. As GBSSI gene expression increases, it actively elongates the
326
short glucan chains; hence their number decreases and the proportion of long glucan chains
327
increases. In addition, 11 enzymes associated with starch synthesis, particularly GBSSII and
328
soluble SS, have been shown to have expressed at the very beginning of wheat grain
329
development (Nadaud et al., 2010), which could be the contributing factor for amylopectin
330
synthesis in the absence of GBSSI, in the completely waxy genotype. Starch biosynthetic
331
enzymes activity assays during endosperm development could provide a better understanding of
332
amylopectin structure formation during grain development, as well as the precise function of
333
GBSSI.
334
Subtle differences were observed in the amylopectin glucan chain length distribution with the
335
reduction of waxy protein dosage during wheat grain development (Fig. 6). Compared to non-
Ac ce p
te
d
M
an
us
cr
ip t
314
21
Page 21 of 33
ip t cr us an M d te Ac ce p 336 337
Fig. 6 22
Page 22 of 33
338
waxy starch, completely waxy starch at 3 DPA showed a higher proportion of short glucan
340
chains of DP 6-8 (Fig. 6), suggesting a relatively earlier synthesis of amylopectin chains in the
341
former as compared to the latter, concurring with a previous report (Cao et al., 2012). At 3 DPA,
342
on DP 6-8 glucan chains, the influence of Wx-B, Wx-D alone or together was the highest,
343
whereas, the influence of Wx-A was the lowest. Similarly at 6 DPA, Wx-B, D affected the
344
glucan short chain fraction more than the other partial waxy starches. This suggests that Wx-B
345
and Wx-D are playing a greater role in amylopectin synthesis during the initial stages of grain
346
development. At 9 DPA, however, Wx-A, B showed a higher proportion of DP 6-15 glucan
347
chains followed by Wx-B and Wx-D. However, Wx-A alone was the least effective in
348
influencing DP 6-8 chains, suggesting that Wx-A only in combination with Wx-B or Wx-D has a
349
stronger effect on the short glucan chains formation. Furthermore, at 12 DPA, Wx-A, B
350
displayed a considerably higher proportion of short chains of DP 6-8 than the other partial waxy
351
and non-waxy wheat starches, suggesting a positive influence of Wx-A on the short chain
352
phenotype during grain filling. At 18 and 24 DPA, Wx-D showed an increase in DP 6-8 glucan
353
chains relative to the non-waxy starch. In addition at 24 and 30 DPA, glucan chains of DP > 20
354
were negatively affected by Wx-A (Fig. 6). This suggests that Wx-B and Wx-D play a stronger
355
role in influencing mid-length glucan chains than Wx-A, during later stages of grain
356
development. The above observations suggest that different waxy isoproteins have different
357
capacities for amylopectin structure formation during wheat grain development; Wx-B and Wx-
358
D being the most effective during early grain development, while Wx-A being active as grain
359
filling starts, followed by Wx-D during grain maturation.
360
4. Conclusions
Ac ce p
te
d
M
an
us
cr
ip t
339
23
Page 23 of 33
Endosperm development is a complex process involving numerous physiological and
362
biochemical transitions. Starch granule morphology continuously changed with each phase of
363
development, resulting in consistent starch and amylose accumulation in non-waxy and partial
364
waxy genotypes. Changes in amylopectin structure were observed until 12 DPA, suggesting the
365
formation of a basic amylopectin skeleton by the grain filling stage. Various waxy isoproteins
366
affected amylopectin structure differently during wheat grain development. In the early stages,
367
Wx-B and Wx-D were the most effective, followed by Wx-A during the mid grain-fill stage and
368
Wx-D during grain maturation. Thus, genome specific GBSSI in wheat considerably affects
369
starch accumulation and glucan chain length distribution during grain development.
an
us
cr
ip t
361
M
370
ACKNOWLEDGEMENTS
372
Canada Research Chairs program, Natural Sciences and Engineering Research Council Canada
373
(Discovery grants program), Canada Foundation for Innovation are acknowledged for financial
374
support. GA is a grateful recipient of the Department of Plant Sciences Devolved Graduate
375
Scholarship.
Ac ce p
te
d
371
24
Page 24 of 33
376
REFERENCES
378
AACC International. Approved Methods of Analysis, 11th Ed. Method 76-13.01. Total starch
379
assay procedure (Megazyme amyloglucosidase/alpha-amylase method). St. Paul, MN: AACC
380
International
381
Abdel-Aal, E.-S.M., Hucl, P., Chibbar, R. N., Han, H. L., & Demeke, T. (2002).
382
Physicochemical and structural characteristics of flours and starches from waxy and nonwaxy
383
wheats. Cereal Chemistry, 79, 458-464.
384
Asare, E. K., Jaiswal, S., Maley, J., Båga, M., Sammynaiken, R., Rossnagel, B. G., & Chibbar,
385
R. N. (2011). Barley grain constituents, starch composition, and structure affect starch in vitro
386
enzymatic hydrolysis. Journal of Agricultural and Food Chemistry, 59, 4743-4754.
387
Bechtel, D., Zayas, I., Kaleikau, L., & Pomeranz, Y. (1990). Size-distribution of wheat-starch
388
granules during endosperm development. Cereal Chemistry, 67, 59-63.
389
Bechtel, D. B., & Wilson, J. D. (2003). Amyloplast formation and starch granule development in
390
hard red winter wheat. Cereal Chemistry, 80, 175-183.
391
Buléon, A., Colonna, P., Planchot, V., & Ball, S. (1998). Starch granules: Structure and
392
biosynthesis. International Journal of Biological Macromolecules, 23, 85-112.
393
Cao, Y-N., Hu, W-G., & Wang, C-S. (2012). Relationship among the key enzymatic activities
394
involved in starch synthesis and amylopectin chain distributions in developing wheat grain.
395
African Journal of Biotechnology, 11, 805-814.
Ac ce p
te
d
M
an
us
cr
ip t
377
25
Page 25 of 33
Chojecki, A. J. S., Bayliss, M. W., & Gale, M. D. (1986). Cell production and DNA
397
accumulation in the wheat endosperm, and their association with grain weight. Annals of Botany,
398
58, 809-817.
399
Dai, Z. (2010). Activities of enzymes involved in starch synthesis in wheat grains differing in
400
starch content. Russian Journal of Plant Physiology, 57, 74-78.
401
Delrue, B., Fontaine, T., Routier, F., Decq, A., Wieruszeski, J-M., Koornhuyse, N., Maddelein,
402
M-L., Fournet, B., & Ball, S. (1992). Waxy Chlamydomonas reinhardtii - monocellular algal
403
mutants defective in amylose biosynthesis and granule-bound starch synthase activity
404
accumulate a structurally modified amylopectin. Journal of Bacteriology, 174, 3612-3620.
405
Demeke, T., Hucl, P., Nair, R., Nakamura, T., & Chibbar, R. N. (1997). Evaluation of Canadian
406
and other wheats for waxy proteins. Cereal Chemistry, 74, 442-444.
407
Demeke, T., Hucl, P., Abdel-Aal, E., Båga, M., & Chibbar, R. N. (1999). Biochemical
408
characterization of the wheat waxy A protein and its effect on starch properties. Cereal
409
Chemistry, 76, 694-698.
410
Denyer, K., Clarke, B., Hylton, C., Tatge, H., & Smith, A. (1996). The elongation of amylose
411
and amylopectin chains in isolated starch granules. Plant Journal, 10, 1135-1143.
412
Dian, W., Jiang, H., & Wu, P. (2005). Evolution and expression analysis of starch synthase III
413
and IV in rice. Journal of Experimental Botany, 56, 623-632.
414
Evers, A. D. (1971). Scanning electron microscopy of wheat starch. III. Granule development in
415
the endosperm. Starch/Stärke, 23, 157-160.
Ac ce p
te
d
M
an
us
cr
ip t
396
26
Page 26 of 33
Ganeshan, S., Drinkwater, J. M., Repellin, A., & Chibbar, R. N. (2010). Selected carbohydrate
417
metabolism genes show coincident expression peaks in grains of in vitro-cultured immature
418
spikes of wheat (Triticum aestivum L.). Journal of Agricultural and Food Chemistry, 58, 4193-
419
4201.
420
Gao, M., & Chibbar, R. N. (2000). Isolation, characterization, and expression analysis of starch
421
synthase IIa cDNA from wheat (Triticum aestivum L.). Genome, 43, 768-775.
422
Gromova, I., & Celis, J. E. (2006). Protein detection in gels by silver staining: A procedure
423
compatible with mass-spectroscopy. In J. E. Celis, N. Carter, T. Hunter, K. Simons, J. V. Small,
424
& D. Shotton (Eds.), Cell Biology: A Laboratory Handbook (pp. 1-8). Amsterdam, Boston:
425
Elsevier Academic Press.
426
Guillon, F., Larŕe, C., Petipas, F., Berger, A., Moussawi, J., Rogniaux, H., Santoni, A., Saulnier,
427
L., Jamme, F., Miquel, M., Lepiniec, L., & Dubreucq, B. (2011). A comprehensive overview of
428
grain development in Brachypodium distachyon variety Bd21. Journal of Experimental Botany,
429
63, 739-755.
430
Hizukuri, S. (1986). Polymodal distribution of the chain lengths of amylopectins, and its
431
significance. Carbohydrate Research, 147, 342-347.
432
Hucl, P., & Chibbar, R. N. (1996). Variation for starch concentration in spring wheat and its
433
repeatability relative to protein concentration. Cereal Chemistry, 73, 756-758.
434
Jane, J., Chen, Y. Y., Lee, L. F., McPherson, A. E., Wong, K. S., Radosavljevic, M. &
435
Kasemsuwan, T. (1999). Effects of amylopectin branch chain length and amylose content on the
436
gelatinization and pasting properties of starch. Cereal Chemistry, 76, 629-637.
Ac ce p
te
d
M
an
us
cr
ip t
416
27
Page 27 of 33
Jobling, S. (2004). Improving starch for food and industrial applications. Current Opinion in
438
Plant Biology, 7, 210-218.
439
Kalinga, D. N., Bertoft, E., Tetlow, I., Liu, Q., Yada, R. Y., & Seetharaman, K. (2014)
440
Formation of amylopectin structure in developing wheat endosperm. Carbohydrate Polymers,
441
112, 325-333.
442
Laudencia-Chingcuanco, D. L., Stamova, B. S., You, F. M., Lazo, G. R., Beckles, D. M., &
443
Anderson, O. D. (2007). Transcriptional profiling of wheat caryopsis development using cDNA
444
microarrays. Plant Molecular Biology, 63, 651-668.
445
Li, C-Y, Feng, C-N., Wang, Y-L., Zhang, R., Gua, W-S., Zhu, X-K., & Peng, Y-X. (2007).
446
Chain length distribution of debranched amylopectin and its relationship with physicochemical
447
properties of starch in different wheat cultivars. Acta Agronomica Sinica, 33, 1240-1245.
448
Li, L., Blanco, M., & Jane, J-L. (2007). Physicochemical properties of endosperm and pericarp
449
starches during maize development. Carbohydrate Polymers, 67, 630-639.
450
Liu, Q., Weber, E., Currie, V., & Yada, R. (2003). Physicochemical properties of starches during
451
potato growth. Carbohydrate Polymers, 51, 213-221.
452
Manners, D. J. (1989). Recent developments in our understanding of amylopectin structure.
453
Carbohydrate Polymers, 11, 87-112.
454
McCleary, B., Gibson, T., & Mugford, D. (1997). Measurement of total starch in cereal products
455
by amyloglucosidase-alpha-amylase method: Collaborative study. Journal of AOAC
456
International, 80, 571-579.
Ac ce p
te
d
M
an
us
cr
ip t
437
28
Page 28 of 33
McDonald, A. M. L., Stark, J. R., Morrison, W. R., & Ellis, R. P. (1991). The composition of
458
starch granules from developing barley genotypes. Journal of Cereal Science, 13, 93-112.
459
Nadaud, I., Girousse, C., Debiton, C., Chambon, C., Bouzidi, M. F., Martre, P., & Branlard, G.
460
(2010). Proteomic and morphological analysis of early stages of wheat grain development.
461
Proteomics, 10, 2901-2910.
462
Olsen, O-A. (2001). Endosperm development: Cellularization and cell fate specification. Annual
463
Review of Plant Physiology and Plant Molecular Biology, 52, 233-267.
464
Olsen, O-A. (2004). Nuclear endosperm development in cereals and Arabidopsis thaliana. Plant
465
Cell, 16, S214-S227.
466
O'Shea, M., Samuel, M., Konik, C., & Morell, M. (1998). Fluorophore-assisted carbohydrate
467
electrophoresis (FACE) of oligosaccharides: Efficiency of labelling and high-resolution
468
separation. Carbohydrate Research, 307, 1-12.
469
Peng, M., Gao, M., Abdel-Aal, E., Hucl, P., & Chibbar, R. N. (1999) Separation and
470
characterization of A- and B-type starch granules in wheat endosperm. Cereal Chemistry, 76,
471
375-379.
472
Peng, M., Hucl, P., & Chibbar, R.N. (2001) Isolation, characterization and expression analysis of
473
starch synthase I from wheat (Triticum aestivum L.). Plant Science 161, 1055–1062.
474
Raeker, M., Gaines, C., Finney, P., & Donelson, T. (1998). Granule size distribution and
475
chemical composition of starches from 12 soft wheat cultivars. Cereal Chemistry, 75, 721-728.
Ac ce p
te
d
M
an
us
cr
ip t
457
29
Page 29 of 33
Ral, J-P., Colleoni, C., Wattebled, F., Dauvillée, D., Nempont, C., Deschamps, P., Li, Z., Morell,
477
M. K., Chibbar, R. N., Purton, S., d’Hulst, C., & Ball, S. (2006). Circadian clock regulation of
478
starch metabolism establishes GBSSI as a major contributor to amylopectin synthesis in
479
Chlamydomonas reinhardtii. Plant Physiology, 142, 305-317.
480
Sharma, R., Sissons, M. J., Rathjen, A. J., & Jenner, C. F. (2002). The null-4A allele at the waxy
481
locus in durum wheat affects pasta cooking quality. Journal of Cereal Science, 35, 287-297.
482
Simmonds, D., & O’Brien, T. (1981). Morphological and biochemical development of the wheat
483
endosperm. In Y. Pomeranz (Ed.), Advances in Cereal Science and Technology Vol. 4 (pp. 5 –
484
70). St. Paul, MN: American Association of Cereal Chemists.
485
Tetlow, I. J., Beisel, K. G., Cameron, S., Makhmoudova, A., Liu, F., Bresolin, N. S., Wait, R.,
486
Morell, M. K., & Emes, M. J. (2008). Analysis of protein complexes in wheat amyloplasts
487
reveals functional interactions among starch biosynthetic enzymes. Plant Physiology, 146, 1878-
488
1891.
489
Vrinten, P. L., & Nakamura, T. (2000). Wheat granule-bound starch synthase I and II are
490
encoded by separate genes that are expressed in different tissues. Plant Physiology, 122, 255-
491
263.
492
Waduge, R. N., Xu, S., Bertoft, E., & Seetharaman, K. (2013) Exploring the surface morphology
493
of developing wheat starch granules by using Atomic Force Microscopy. Starch/ Stärke, 65, 398-
494
409.
Ac ce p
te
d
M
an
us
cr
ip t
476
30
Page 30 of 33
Yasui, T., Matsuki, J., Sasaki, T., & Yamamori, M. (1996). Amylose and lipid contents,
496
amylopectin structure, and gelatinisation properties of waxy wheat (Triticum aestivum) starch.
497
Journal of Cereal Science, 24, 131-137.
498
Zhao, X. C., Batey, I. L., Sharp, P. J., Crosbie, G., Barclay, I., Wilson, R., Morell, M. K., &
499
Appels, R. (1998). A single genetic locus associated with starch granule properties and noodle
500
quality in wheat. Journal of Cereal Science, 27, 7-13.
Ac ce p
te
d
M
an
us
cr
ip t
495
31
Page 31 of 33
501
FIGURE CAPTIONS
503
Fig. 1 SEM images of starch granules isolated at different development stages in non-waxy (A)
504
and completely waxy (B) genotypes.
505
Fig. 2 Total starch concentration in near isogenic (waxy locus) lines during wheat grain
506
development.
507
Fig. 3 Amylose concentration in near isogenic (waxy locus) lines during wheat grain
508
development.
509
Fig. 4 Denaturing gel electrophoresis analysis showing the presence of starch granule-bound
510
proteins (silver staining) during grain development (3 to 30 DPA) in non-waxy (A) and
511
completely waxy (B) genotypes.
512
Fig. 5 Amylopectin chain length distribution during wheat grain development (3 – 30 DPA) in a
513
non-waxy genotype.
514
Fig. 6 Amylopectin chain length distribution in partial and completely waxy wheat genotypes
515
during grain development (3, 6, 9, 12, 18, 24, and 30 DPA). Subtractive graphs depict relative
516
area in a particular waxy-null genotype subtracted from that of normal genotype. Values are
517
based on average of three biological replicates.
Ac ce p
te
d
M
an
us
cr
ip t
502
32
Page 32 of 33
519
520
521
Highlights
ip t
522
1. Wheat GBSSI ( (Wx) dosage influence starch accumulation during grain development
524
2. Wx-A influences starch accumulation at early stages of grain development
525
3. Wx-B and Wx-D affects synthesis of amylopectin mid-length glucan chains
526
4. Waxy starch amylopectin has higher fibril height than non-waxy
us
cr
523
Ac ce p
te
d
M
an
527
34
Page 33 of 33
